AutoML Leaderboard

Best model	name	model_type	metric_type	metric_value	train_time
	1_Baseline	Baseline	logloss	0.628023	10.83
	2_DecisionTree	Decision Tree	logloss	1.51356	60.51
	3_Linear	Linear	logloss	0.361632	16.68
	4_Default_Xgboost	Xgboost	logloss	0.499665	17.22
	5_Default_NeuralNetwork	Neural Network	logloss	0.454204	10.87
	6_Default_RandomForest	Random Forest	logloss	0.350767	24.25
the best	Ensemble	Ensemble	logloss	0.328403	4.29

AutoML Performance

AutoML Performance Boxplot

Features Importance

features importance across models

Spearman Correlation of Models

models spearman correlation

Summary of 1_Baseline

<< Go back

Baseline Classifier (Baseline)

n_jobs: -1
explain_level: 2

Validation

validation_type: split
train_ratio: 0.75
shuffle: True
stratify: True

Optimized metric

logloss

Training time

7.7 seconds

Metric details

	score	threshold
logloss	0.628023	nan
auc	0.5	nan
f1	0.486486	0.283929
accuracy	0.321429	0.283929
precision	0.321429	0.283929
recall	1	0.283929
mcc	0	0.283929

Metric details with threshold from accuracy metric

	score	threshold
logloss	0.628023	nan
auc	0.5	nan
f1	0.486486	0.283929
accuracy	0.321429	0.283929
precision	0.321429	0.283929
recall	1	0.283929
mcc	0	0.283929

Confusion matrix (at threshold=0.283929)

	Predicted as 0	Predicted as 1
Labeled as 0	0	38
Labeled as 1	0	18

Learning curves

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Kolmogorov-Smirnov Statistic

Precision-Recall Curve

Calibration Curve

Cumulative Gains Curve

Lift Curve

<< Go back

Summary of 2_DecisionTree

<< Go back

Decision Tree

n_jobs: -1
criterion: gini
max_depth: 3
explain_level: 2

Validation

validation_type: split
train_ratio: 0.75
shuffle: True
stratify: True

Optimized metric

logloss

Training time

56.9 seconds

Metric details

	score	threshold
logloss	1.51356	nan
auc	0.729532	nan
f1	0.684211	0.297472
accuracy	0.785714	0.297472
precision	0.714286	0.8125
recall	0.833333	0
mcc	0.524389	0.297472

Metric details with threshold from accuracy metric

	score	threshold
logloss	1.51356	nan
auc	0.729532	nan
f1	0.684211	0.297472
accuracy	0.785714	0.297472
precision	0.65	0.297472
recall	0.722222	0.297472
mcc	0.524389	0.297472

Confusion matrix (at threshold=0.297472)

	Predicted as 0	Predicted as 1
Labeled as 0	31	7
Labeled as 1	5	13

Learning curves

Permutation-based Importance

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Kolmogorov-Smirnov Statistic

Precision-Recall Curve

Calibration Curve

Cumulative Gains Curve

Lift Curve

SHAP Importance

SHAP Dependence plots

Dependence (Fold 1)

SHAP Dependence from Fold 1

SHAP Decision plots

Top-10 Worst decisions for class 0 (Fold 1)

SHAP worst decisions class 0 from Fold 1

Top-10 Best decisions for class 0 (Fold 1)

SHAP best decisions class 0 from Fold 1

Top-10 Worst decisions for class 1 (Fold 1)

SHAP worst decisions class 1 from Fold 1

Top-10 Best decisions for class 1 (Fold 1)

SHAP best decisions class 1 from Fold 1

<< Go back

Summary of 3_Linear

<< Go back

Logistic Regression (Linear)

n_jobs: -1
explain_level: 2

Validation

validation_type: split
train_ratio: 0.75
shuffle: True
stratify: True

Optimized metric

logloss

Training time

12.3 seconds

Metric details

	score	threshold
logloss	0.361632	nan
auc	0.903509	nan
f1	0.810811	0.364006
accuracy	0.875	0.364006
precision	1	0.757
recall	1	0.000133374
mcc	0.718164	0.364006

Metric details with threshold from accuracy metric

	score	threshold
logloss	0.361632	nan
auc	0.903509	nan
f1	0.810811	0.364006
accuracy	0.875	0.364006
precision	0.789474	0.364006
recall	0.833333	0.364006
mcc	0.718164	0.364006

Confusion matrix (at threshold=0.364006)

	Predicted as 0	Predicted as 1
Labeled as 0	34	4
Labeled as 1	3	15

Learning curves

Coefficients

feature	Learner_1
serum_creatinine	0.459312
age	0.414471
diabetes	0.353401
high_blood_pressure	0.326843
anaemia	0.297311
creatinine_phosphokinase	0.159132
smoking	-0.0403247
sex	-0.159142
ejection_fraction	-0.478172
platelets	-0.520779
serum_sodium	-0.524161
time	-1.45222
intercept	-1.70399

Permutation-based Importance

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Kolmogorov-Smirnov Statistic

Precision-Recall Curve

Calibration Curve

Cumulative Gains Curve

Lift Curve

SHAP Importance

SHAP Dependence plots

Dependence (Fold 1)

SHAP Dependence from Fold 1

SHAP Decision plots

Top-10 Worst decisions for class 0 (Fold 1)

SHAP worst decisions class 0 from Fold 1

Top-10 Best decisions for class 0 (Fold 1)

SHAP best decisions class 0 from Fold 1

Top-10 Worst decisions for class 1 (Fold 1)

SHAP worst decisions class 1 from Fold 1

Top-10 Best decisions for class 1 (Fold 1)

SHAP best decisions class 1 from Fold 1

<< Go back

Summary of 4_Default_Xgboost

<< Go back

Extreme Gradient Boosting (Xgboost)

n_jobs: -1
objective: binary:logistic
eta: 0.075
max_depth: 6
min_child_weight: 1
subsample: 1.0
colsample_bytree: 1.0
eval_metric: logloss
explain_level: 2

Validation

validation_type: split
train_ratio: 0.75
shuffle: True
stratify: True

Optimized metric

logloss

Training time

12.9 seconds

Metric details

	score	threshold
logloss	0.499665	nan
auc	0.83114	nan
f1	0.723404	0.247828
accuracy	0.785714	0.444885
precision	0.666667	0.799324
recall	1	0.0924742
mcc	0.58757	0.247828

Metric details with threshold from accuracy metric

	score	threshold
logloss	0.499665	nan
auc	0.83114	nan
f1	0.684211	0.444885
accuracy	0.785714	0.444885
precision	0.65	0.444885
recall	0.722222	0.444885
mcc	0.524389	0.444885

Confusion matrix (at threshold=0.444885)

	Predicted as 0	Predicted as 1
Labeled as 0	31	7
Labeled as 1	5	13

Learning curves

Permutation-based Importance

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Kolmogorov-Smirnov Statistic

Precision-Recall Curve

Calibration Curve

Cumulative Gains Curve

Lift Curve

SHAP Importance

SHAP Dependence plots

Dependence (Fold 1)

SHAP Dependence from Fold 1

SHAP Decision plots

Top-10 Worst decisions for class 0 (Fold 1)

SHAP worst decisions class 0 from Fold 1

Top-10 Best decisions for class 0 (Fold 1)

SHAP best decisions class 0 from Fold 1

Top-10 Worst decisions for class 1 (Fold 1)

Top-10 Best decisions for class 1 (Fold 1)

SHAP best decisions class 1 from Fold 1

<< Go back

Summary of 5_Default_NeuralNetwork

<< Go back

Neural Network

n_jobs: -1
dense_1_size: 32
dense_2_size: 16
learning_rate: 0.05
explain_level: 2

Validation

validation_type: split
train_ratio: 0.75
shuffle: True
stratify: True

Optimized metric

logloss

Training time

4.5 seconds

Metric details

	score	threshold
logloss	0.454204	nan
auc	0.831871	nan
f1	0.709677	0.47667
accuracy	0.839286	0.47667
precision	0.888889	0.676714
recall	1	0.00514851
mcc	0.617774	0.47667

Metric details with threshold from accuracy metric

	score	threshold
logloss	0.454204	nan
auc	0.831871	nan
f1	0.709677	0.47667
accuracy	0.839286	0.47667
precision	0.846154	0.47667
recall	0.611111	0.47667
mcc	0.617774	0.47667

Confusion matrix (at threshold=0.47667)

	Predicted as 0	Predicted as 1
Labeled as 0	36	2
Labeled as 1	7	11

Learning curves

Permutation-based Importance

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Kolmogorov-Smirnov Statistic

Precision-Recall Curve

Calibration Curve

Cumulative Gains Curve

Lift Curve

<< Go back

Summary of 6_Default_RandomForest

<< Go back

Random Forest

n_jobs: -1
criterion: gini
max_features: 0.9
min_samples_split: 30
max_depth: 4
eval_metric_name: logloss
explain_level: 2

Validation

validation_type: split
train_ratio: 0.75
shuffle: True
stratify: True

Optimized metric

logloss

Training time

19.8 seconds

Metric details

	score	threshold
logloss	0.350767	nan
auc	0.912281	nan
f1	0.818182	0.288106
accuracy	0.857143	0.288106
precision	0.818182	0.723054
recall	1	0.0200628
mcc	0.739296	0.288106

Metric details with threshold from accuracy metric

	score	threshold
logloss	0.350767	nan
auc	0.912281	nan
f1	0.818182	0.288106
accuracy	0.857143	0.288106
precision	0.692308	0.288106
recall	1	0.288106
mcc	0.739296	0.288106

Confusion matrix (at threshold=0.288106)

	Predicted as 0	Predicted as 1
Labeled as 0	30	8
Labeled as 1	0	18

Learning curves

Permutation-based Importance

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Kolmogorov-Smirnov Statistic

Precision-Recall Curve

Calibration Curve

Cumulative Gains Curve

Lift Curve

SHAP Importance

SHAP Dependence plots

Dependence (Fold 1)

SHAP Dependence from Fold 1

SHAP Decision plots

Top-10 Worst decisions for class 0 (Fold 1)

SHAP worst decisions class 0 from Fold 1

Top-10 Best decisions for class 0 (Fold 1)

SHAP best decisions class 0 from Fold 1

Top-10 Worst decisions for class 1 (Fold 1)

SHAP worst decisions class 1 from Fold 1

Top-10 Best decisions for class 1 (Fold 1)

SHAP best decisions class 1 from Fold 1

<< Go back

Summary of Ensemble

<< Go back

Ensemble structure

Model	Weight
3_Linear	2
6_Default_RandomForest	3

Metric details

	score	threshold
logloss	0.328403	nan
auc	0.932749	nan
f1	0.829268	0.357868
accuracy	0.875	0.357868
precision	0.909091	0.604225
recall	1	0.012091
mcc	0.746678	0.357868

Metric details with threshold from accuracy metric

	score	threshold
logloss	0.328403	nan
auc	0.932749	nan
f1	0.829268	0.357868
accuracy	0.875	0.357868
precision	0.73913	0.357868
recall	0.944444	0.357868
mcc	0.746678	0.357868

Confusion matrix (at threshold=0.357868)

	Predicted as 0	Predicted as 1
Labeled as 0	32	6
Labeled as 1	1	17